“We don’t just make a product for the growers, we do it together with them”

“In a vertical farm, energy consumption is one of the biggest operating costs, and a large part of it comes from artificial lighting. It is important to reduce these costs if we want to make Controlled Environment Agriculture (CEA) a profitable cultivation method and to is the biggest problem we’re trying to solve with our technology,” said Dr. Rakibul Islam, a biomedical researcher at Photosynthetic, a subsidiary of Rift Labs.

He explains that the Photosynthetic R&D lab provides a platform to conduct scientific tests by comparing environmental variables while monitoring plant growth. It provides valuable insight to reduce resource wastage, for example by adjusting photoperiod, wavelength, temperature, etc. for optimal plant growth. It gives vertical farms a recipe for energy-efficient plant growth that they can use in the growing environment.

Rakibul Islam

As a researcher, Rakibul has always had a huge interest in artificial intelligence. He has worked in a large AI research group where they developed and evaluated the use of artificial intelligence in cancer diagnosis and prognosis. At Photosynthetic, however, he is tasked with innovating AI solutions for growers.

“It is very satisfying to use the possibilities of my AI knowledge to make CEA more efficient by creating intelligent processes and workflows. This project combines both my passion for growing plants and my intellectual curiosity.”

Grow with the grower
Photosynthetic’s expertise lies in generating data and precisely controlling inputs to plant growth using proprietary software and patented technology to compose light at different wavelengths. For example, the autonomous R&D laboratory allows producers to better understand the environmental needs of a candidate crop and experiment for efficiency without compromising the plant’s physiology.

By monitoring the environmental conditions in this lab, one can mimic the “problem situation” of the growers and generate highly relevant data needed to train machine learning models. The stored optimal growth data can be used to simulate growth and predict crop harvest time with computers.

“We don’t just make a product for the growers, we do it together with them. With Photosynthetic we want to establish ourselves as a research partner and technology supplier for growers all over the world. We have a data-driven approach where we want to leverage our expertise in science about light and deliver smart solutions that enable our customers to continuously improve their production process,” says Rakibul.

AI explained
As AI is a large field, this term is usually used interchangeably with Machine Learning (ML), Deep Learning (DL), etc. It consists of a set of tools that enable computers to learn without being explicitly programmed, explains Rakibul .

“Traditional software is rule-based, where exact instructions are given to a computer to solve a problem, meaning we need input and rules to get an output. In machine learning, we need input and previous output to get a model so that it can predict the output when it receives a new input. Machine learning applications are all around us today, the most important thing is that they are useful,” says Rakibul.

This is how the machine learning model works to unlock your phone using a facial recognition system. Spam filters in your email, search engine results and the movie recommendation system on Netflix; so are machine learning models. There are applications of AI in flying drones and talking robots, so it’s not entirely wrong to think of that when you think of the term AI.

“We should design productivity technology not just to make sure it sounds good, but also that it adds value to users by reducing unnecessary tasks and complexity. The best AI products and services are often invisible and seamlessly integrated into the workflow.”

The laboratory

Current role of AI in CEA
According to Rakibul, CEA is one of the ideal candidates to harness the power of AI. In fact, CEA facilities are designed to monitor and manage growing conditions to improve the sustainability of horticultural production. Growers make their decisions based on data for better use of resources and higher productivity, quality, profitability and sustainability.

AI can play a role in these areas; optimizing electricity consumption, one of the biggest cost drivers in CEA plants; process automation to increase productivity; optimizing plant growth and yield forecasting to reduce uncertainty; detect bad products to ensure quality.

“The technological capacity is there. But there are no ready-made solutions. These solutions are data-dependent and must therefore be developed in collaboration with the growers, based on the type of optimization they need,” he states.

Improving farming with AI
Indeed, AI solutions for indoor plant cultivation require a remarkably diverse set of expertise. Of course, coding a machine learning model is a critical part, but not the only one, nor is it the biggest challenge. The development of artificial intelligence requires many different aspects of a complex infrastructure such as data collection, feature extraction, supporting infrastructure, monitoring, etc.

“However, I see AI as one of the many interesting tools we can use to increase yields and reduce growth costs. In my opinion, the possibilities are endless and there is still much to explore. We are working with AI through our multidisciplinary team, consisting of plant scientists, AI product managers and mechatronics, electronics and software engineers, to support our customers in defining their challenges and seeing them as problems that can be solved with machine learning.Routine manual tasks are reduced, decision making is supported and ultimately end the production process is optimized.”

For more information:
Øyvind Hasund Dahl
Photosynthetic (by Rift Labs)

Leave a Comment